IMMUNOGLOBULIN A (IGA), IGAL, AND IGA2 ANTIBODIES TO CANDIDA ALBICANS IN WHOLE AND PAROTID SALIVA IN HUMAN
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
INFECTION AND IMMUNITY, Mar. 1994, p. 892-896 Vol. 62, No. 3 0019-9567/94/$04.00+0 Copyright C) 1994, American Society for Microbiology Immunoglobulin A (IgA), IgAl, and IgA2 Antibodies to Candida albicans in Whole and Parotid Saliva in Human Immunodeficiency Virus Infection and AIDS MAEVE M. COOGAN,t SIMON P. SWEET, AND STEPHEN J. CHALLACOMBE* Centre for the Study of the Oral Manifestations of HIV Infection, UMDS Guy's Hospital, London, SE1 9RT, United Kingdom Received 1 June 1993/Returned for modification 13 August 1993/Accepted 21 December 1993 Human immunodeficiency virus (HIV)-infected individuals are predisposed to recurrent oral candidiasis, Downloaded from http://iai.asm.org/ on October 6, 2015 by guest and, although it has been assumed that this is because of deficient mucosal immune responses, this has not been properly established. The present study aimed to compare the concentrations and secretion rates of immunoglobulin A (IgA) and IgA subclass antibodies to Candida albicans in whole and parotid saliva samples from HIV-infected patients, AIDS patients, and control subjects. Levels of IgA antibody to Candida species in whole saliva were higher in the HIV group than in the controls and were highest in the AIDS group (P < 0.05). In parotid saliva, the mean antibody levels were significantly greater in HIV-positive patients than in controls (P < 0.05) but fell to lower levels in the AIDS group. The secretion rates of Candida antibodies in parotid saliva were reduced in AIDS patients compared with HIV patients. The specific activities of the IgA antibodies and both subclasses were significantly higher in the HIV and AIDS patients than in the controls in both whole and parotid saliva (P < 0.05). Antibody levels were significantly correlated with the numbers of Candida organisms isolated from saliva (P < 0.05). These results suggest clear differences in salivary antibody profiles among HIV-infected, AIDS, and control subjects and are indicative of a response to antigenic challenge by infecting Candida species. No obvious defect in the mucosal immune response in the HIV or AIDS groups that might account for the increased prevalence of candidiasis was apparent. The main sequelae of human immunodeficiency virus (HIV) than control subjects. Jackson (10) also reported that IgA2 was infection are frequent and persistent opportunistic secondary significantly reduced in whole saliva from AIDS patients infections at mucosal surfaces. Oral candidal infections are the compared with controls. These observations support the sug- most common mucosal manifestation of HIV infection and gestion that the level of secretory IgA may contribute to the may occur in up to 50% of HIV-infected subjects (13) and 90% frequent mucosa-related opportunistic infections seen in these of AIDS patients (13, 15, 19). This suggests that the oral patients. mucosal defense mechanisms of these patients are compro- There are few reports of specific antibodies in relation to mised. Immunological protection of mucosal surfaces is medi- mucosal immunity in HIV infection. A study of 10 HIV- ated primarily by the secretory immune system, particularly infected subjects found that levels of IgA antibody to C. immunoglobulin A (IgA) in secretions, which affords protec- albicans in whole saliva were raised but found no correlation tion by inhibiting adherence and penetration of microorgan- between candidal carriage and antibody titers (23). Another isms and foreign proteins to mucosal tissues (6, 22, 25, 26). preliminary study reported similar increases in C. albicans IgA Secretory IgA may play a key role in protection against oral antibody levels in whole saliva from HIV-infected patients but candidiasis, since IgA molecules have been identified on the noted decreased levels in patients with candidiasis (27). These surfaces of the Candida albicans cells (8), and it has been somewhat contradictory studies suggest that alterations in IgA shown that secretory IgA antibodies can inhibit the adherence responses may occur in association with HIV infection; how- of C. albicans to epithelial cells (5, 24). Decreased production ever, the nature of this association and the role of specific of IgA might therefore favor oral colonization and the devel- antibody in protection of mucosal surfaces remain to be fully opment of candidiasis. elucidated. Therefore, the aims of the present study were to HIV infects cells via interactions with CD4 receptor sites on analyze the salivary concentrations and secretion rates of IgA a number of human cell types, including the helper/inducer and the subclasses IgAl and IgA2 to C. albicans in whole and subset of T-lymphocytes (7) which play a pivotal role in most parotid saliva from HIV-infected, AIDS, and control subjects. immunological processes, including the induction of secretory IgA responses (17). Abnormalities of salivary immunoglobulin levels in patients infected with HIV have been reported. MATERIALS AND METHODS Muller et al. (18) found that patients with AIDS had signifi- cantly lower levels of IgA, IgAl, and IgA2 in parotid saliva Subjects and collection of saliva. The study population consisted of 124 males: 45 asymptomatic HIV antibody-posi- tive subjects without AIDS-defining criteria with a mean age of * Corresponding author. Mailing address: Department of Oral 36 years (standard deviation [SD], 11.1; range 20 to 69), 27 Medicine and Pathology, UMDS Guy's Tower, Guy's Hospital, Lon- AIDS patients with a mean age of 40 years (SD, 10.5; range 24 don SEI 9RT, United Kingdom. Phone: (44) 71-955-4256. Fax: (44) to 64), and 52 male control subjects with a mean age of 30 years 71-407-7361. (SD, 11.9; range 19 to 73). Asymptomatic HIV-infected sub- tPresent address: Department of Oral Pathology, University of the jects who had not received antibiotic or antimycotic therapy for Witwatersrand, Johannesburg, South Africa. at least 3 months were selected, and none of the AIDS subjects 892
VOL. 62, 1994 SALIVARY CANDIDA IgA ANTIBODIES IN HIV INFECTION 893 were on antibiotic or antimycotic therapy at the time of levels. Samples with coefficients of variation in excess of 15% sampling. Whole unstimulated saliva samples were collected were retested. Secretion rates were calculated by multiplying from all subjects by expectoration into sterile plastic universals salivary concentrations by salivary flow rates, and specific for 10 min and were clarified by centrifugation at 11,500 x g activities were calculated by dividing salivary concentrations by for 10 min. Parotid saliva samples were collected from 21 total IgA or IgA subclass concentrations. The data were HIV-infected, 11 AIDS, and 24 control subjects by using analyzed by an unpaired Student's t-test, the Mann-Whitney U Lashley cups (16) placed directly over one parotid duct and test, or the chi-square test as appropriate. with salivary flow stimulated by the application of 100 ,ul of 5% Total IgA, IgAl, and IgA2 concentrations were measured by citric acid to the tongue at 2-min intervals for 10 min. a sandwich ELISA technique. Microtiter plates were coated Saliva samples were cultured for the presence of yeasts by with 100 ,ul of a 1:1,000 dilution of rabbit anti-human IgA plating 100 ,u onto Sabouraud's dextrose agar plates and (Dakopatts A/S, Glostrup, Denmark) in PBS-azide by incuba- incubating at 37°C for 72 h. The CFU of yeasts per ml of saliva tion at 37°C for 2 h. Doubling dilutions of saliva were made were counted, and isolates were identified to species level by from 1:500 to 1:64,000 for IgA and from 1:100 to 1:12,800 for using the API 20 AUX system (BioMerieux, Marcy l'Etoile, IgAl and IgA2. Bound IgA was detected as in the Candida France). ELISA, and salivary IgA concentrations were calculated by Detection of antibodies. Levels of IgA, IgAl, and IgA2 reference to a standard preparation of purified human colos- antibodies to C. albicans NCPF 3153 (serotype A) in saliva tral IgA (Sigma). IgA in both colostrum and saliva has been Downloaded from http://iai.asm.org/ on October 6, 2015 by guest were determined by a solid-phase enzyme-linked immunosor- shown to consist predominantly of secretory IgA, and the bent assay (ELISA) (2). Strain NCPF 3153 was considered calculated concentrations of IgAl and IgA2 were adjusted antigenically representative of C. albicans, since 20 samples of according to the IgAl/IgA2 ratio in colostrum, estimated to be saliva assayed against 100 other strains of C. albicans isolated 53:47 (4, 9, 14). Standard curves with colostrum IgA were from HIV-infected and control subjects gave mean antibody constructed from six serial dilutions, in duplicate, beginning at levels that were not significantly different from those found 160 ng/ml. against NCPF 3153. Candida cells were grown on Sabouraud's Antibody specificity. Antibody specificity was determined by dextrose agar and were suspended overnight in buffered formal adsorption of six saliva samples with formalin-fixed C. albicans saline. The cells were washed three times in phosphate- NCPF 3153 whole cells. A washed suspension of fixed C. buffered saline (PBS) containing 0.0005% sodium azide (PBS- albicans cells in PBS was adjusted to an optical density of azide) and then suspended in PBS containing 0.3% methyl approximately 1.5 at 600 nm, and the cells were centrifuged to glyoxal to an optical density of 1.7 at 540 nm. Immulon 2 form a pellet. The supernatant was replaced by saliva diluted microtiter plates (Dynatech Laboratories, Inc., Chantilly, Va.) 1:4 with PBS, and adsorption was performed at 37°C for 1 h were coated with Candida cells by incubation with 100 RI of and then overnight at 4°C. The cells were removed by centrif- yeast suspension in each well at 37°C for 2 h. The plates were ugation, and the supernatant was assayed for C. albicans washed three times with PBS-azide, and subsequent nonspe- antibodies by ELISA as described. cific binding was blocked by incubation at 37°C for 2 h with 200 [LI of PBS containing 0.5% bovine serum albumin (BSA) and RESULTS 0.05% Tween 20 (T20). Saliva samples were diluted 1:5 in PBS-BSA-T20, dispensed into the top row of each microtiter Antibodies to C. albicans. Concentrations of IgA, IgAl, and plate, and diluted further to 1:640 in eight doubling serial IgA2 antibodies specific to C. albicans in whole saliva were dilutions, leaving 100 RI in each well. Salivary IgA bound greater in both the HIV-infected and the AIDS groups than in during overnight incubation at 4°C was then detected by the control subjects (Fig. la). These differences were statisti- incubation for 2 h at 37°C with 100 [LI of mouse monoclonal cally significant for AIDS patients compared with control IgG anti-human IgA (diluted 1:1,000), IgAl (1:500), or IgA2 subjects for IgA and IgAl, and also for IgA and IgA2 in the (1:500; Becton Dickinson, San Jose, Calif.) and then by HIV-infected subjects (P < 0.05; Fig. la). In parotid saliva, incubation for 1 h at 37°C with 100 ,u of rabbit anti-mouse IgG statistically significant increases in IgA, IgAl, and IgA2 anti- alkaline phosphatase conjugate (1:300; Sigma Chemical Co., body levels were found in HIV-infected and AIDS subjects Poole, England). Phosphatase activity was assessed at room compared with control subjects (P < 0.05; Fig. lb). There was temperature with a 100-,ul p-nitrophenol phosphate substrate no obvious change in the IgAl/IgA2 ratio in HIV-infected or (1 mg/ml; Sigma) dissolved in diethanolamine buffer (pH 9.8). AIDS patients compared with control subjects. The reactions were allowed to proceed until sufficient yellow Adsorption of saliva samples with formalin-fixed C. albicans had developed (20 to 60 min) and were then stopped by the cells reduced the antibody titer by greater than 95%. addition of 50 RI of 3 M NaOH prior to reading of the plates Specific activity. Candida antibody-specific activities (in in a Dynatech automatic ELISA reader at 405 nm. All antibody units per microgram of IgA) were calculated by dividing the solutions were prepared in PBS-BSA-T20. The microtiter Candida IgA antibody levels (units per milliliter) by the total plates were washed three times with PBS between all stages. IgA concentrations (micrograms per milliliter) in saliva. The Salivary antibody concentrations were calculated by refer- specific activities in whole saliva were markedly increased in ence to a pooled standard saliva obtained from 10 healthy both the HIV-infected and AIDS groups compared with the males with high levels of anti-Candida antibody activity. The controls for both of the two subclasses (P < 0.05; Fig. 2a). A standard saliva was assigned an arbitrary value of 1,000 U and similar pattern was found with parotid saliva (Fig. 2b), al- was assayed in duplicate for every microtiter plate. Antibody though the specific activity in the AIDS group, which was concentrations were determined from the mean of all values greater than that in the controls, was less than that in the falling within the range of the standard curve. The eight HIV-infected group. dilutions were also used to compare parallelism between the Secretion rates. Secretion rates are a function of both standard and sample curves. Coefficients of variation for each antibody concentrations and whole or parotid saliva flow rates, sample were calculated by dividing the standard deviations of expressed as the numbers of antibody units of IgA, IgAl, or the antibody concentrations of the dilutions used, determined IgA2 secreted per minute. The mean unstimulated whole by reference to the standard curve, by the mean antibody saliva flow rates for the HIV-infected, AIDS, and control
894 COOGAN ET AL. INFECT. IMMUN. a a 450 00 ~~~~~~~~~~~~~~~~~~~~~00 700 350 600ibody Antibody soogl 500 S 1 * * Antibody secretion 300 250 concentration 400 rt 0 (units/mi)~ 300 ~~ ~ ~ ~ ~ ~~~~~~~~~(ntsmn 5 20070 150 C HA C HA CH ACHA IC iAl CIHIA IgA IgAl IgA2 IgA IgAl IgA2 b 70b10 160. 600- (unts/ml)l> 0 Cl + < tI[tntl 140- Downloaded from http://iai.asm.org/ on October 6, 2015 by guest ~~~~~~~~~~~~~~~~~~Antibody 120- Antibody 400 secretion 100r concentration rate (units/mi) 300' (units/miv /gland) 60 200 100 X 20 (units/0g) 6 Xi x E significntly different fr0C tHoA CH hA C H C HA C HA C HA IgA IgAl IgA2 IgA IgAl IgA2 FIG. 1. Concentrations of IgA, IgAl, and IgA2 antibodies to C. FIG. 3. Anti-Candida IgA, IgAl, and IgA2 antibody secretion rates albicans whole (a) and parotid (b) in saliva from control subjects (C) in whole (a) and parotid (b) saliva from controlsubjects (C), HIV- and HIV-infected (H) and AIDS (A) patients. Error bars refer to the positive subjects (H), and AIDS patients (A). Error bars refer to the SEMs. SEMs. groups were 0.63 (± 0.07), 0.52 (± 0.10), and 0.67 (± 0.04) mi/mmn (± standard error of the mean [SEM]), respectively. The mean secretion rate of antibodies in the AIDS group was significantly greater than that for the control group (P < 0.05; Fig. 3a), even though the salivary flow rates were lower. The a 16man stimulated parotid saliva flow rates for the HI V-netd 14 AIDS, and control groups (±0.06), 0.20 (±0.04), were 0.26 and 0.29 0.03) ml/min SEM), respectively. per gland The mean parotid flow rate of the AIDS group was signifi- Antibody lo0. cantly lower than that of the control group (P < 0.05) and activity }HIV-inresulted in mean antibody secretion rates that were not (unitS4tg) 6 significantly different from those of the controls (Fig. 3b). 4- ~~~~~~~~~~However, the significant increases in IgA and IgA2 antibody levels in the parotid saliva of the HI V-infected group (Fig. lb) 2- ~~~~~~~~~~~were maintained in the secretion rates (P < 0.05; Fig. 3b). CHACandida species in saliva. Thirteen of the 52 control subjects (nt/)412 HA CHA carried C. albicans in their saliva, but only 2 carried more than IgA IgAl IgA2 ~~~~~~~~~~1,500 CFU/ml of saliva. In contrast, 18 of the 44 HI V-infected ~~~~~~~~~~~~patients (P 0.001) and < 11 of the 27 AIDS patients (P < 0.001) had more than 1,500 yeast CFU/ml of whole saliva ~~~(Table 1). All of the yeasts isolated from the control group Antibody ere identified to species level as C. albicans, and although specific (units/p.g) 4-~ TAL .Carriage rates ofyeasts in whole saliva of control, No. (%) of subjects with C H A rICH'AI CHA Subject 1,500 CFU/ml IgA IgAl IgA2 Control (n = 52) 39 (75) 11 (21) 2 (4) FIG. 2. Specific activities of anti-Candida IgA, IgAl, and IgA2 HIV infected (n = 45) 13 (29) 14 (31) 18 (40)- antibodies in whole (a) and parotid (b) saliva from control subjects AIDS (n = 27) 10 (37) 6 (22) 11 (41)- (C), HIV-positive subjects. (H), and AIDS patients (A). Error bars refer to the SEMs. ' Significant difference (P < 0.001) compared with control subjects.
VOL. 62, 1994 SALIVARY CANDIDA IgA ANTIBODIES IN HIV INFECTION 895 TABLE 2. Whole-saliva IgA antibodies to Candida species in This increase in antibody secretion may be related to antigen relation to candidal carriage in HIV-infected and AIDS subjects challenge, since almost half of the HIV-infected and AIDS Salivary Candida count Anti-C. albicans IgA patients carried more than 103 yeast CFU/ml of saliva, and (CFU/ml) (U/ml of saliva + salivary antibody production significantly correlated with sali- SEM) vary Candida load (Table 2). Increased salivary IgA Candida 10 (n = 49) ................................... 860 ± 158a infection, instead of having a protective role. >5,000 (n = 16) ................................... 1,077 ± 376a The correlation between oral infection and salivary response a Significant difference (P < 0.05) compared with
896 COOGAN ET AL. INFECT. IMMUN. other factors contribute to the increased prevalence of candi- A and their subversion by immunoglobulin A proteases. Micro- diasis found in HIV-infected individuals. biol. Rev. 52:296-303. 13. Korting, H. C., M. Ollert, A. Georgii, and M. Froschel. 1989. In vitro susceptibilities and biotypes of Candida albicans isolates ACKNOWLEDGMENTS from oral cavities of persons infected with HIV. J. Clin. Microbiol. This study was supported by the Oral and Dental Research Trust 26:2626-2631. and the Dunhill Medical Trust. 14. Ladjeva, I., J. H. Peterman, and J. Mestecky. 1989. IgA subclass of human colostral antibodies specific for microbial and food anti- REFERENCES gens. Clin. Exp. Immunol. 78:85-90. 15. Langford-Kuntz, A. A., R. Ruchel, and P. A. Reichart. 1988. Oral 1. Arendorf, T. M., and D. M. Walker. 1980. The prevalence and manifestations of candidiasis in HIV-infected patients. Clinical intraoral distribution of Candida albicans. Arch. Oral Biol. 25:1- and microbiological studies. Dtsch. Z. Mund. Kiefer. Gesichtschir. 10. 12:28-35. 2. Challacombe, S. J. 1988. Application of ELISA to microbiology, p. 16. Lashley, K. S. 1916. Reflex secretion of the human parotid gland. 319-342. In D. M. Kemeny and S. J. Challacombe (ed.), ELISA J. Exp. Psychol. 1:462-465. and other solid phase immunoassays: theoretical and practical 17. McGhee, J. R., J. Mestecky, C. 0. Elson, and H. Kiyono. 1989. aspects. John Wiley and Sons, Chichester, United Kingdom. Regulation of IgA synthesis and immune response by T cells and 3. Challacombe, S. J. 1990. Immunology of oral candidosis, p. interleukins. J. Clin. Immunol. 9:175-199. 104-123. In L. P. Samaranayake and T. W. MacFarlane (ed.), Oral Downloaded from http://iai.asm.org/ on October 6, 2015 by guest candidosis. Wright, London. 18. Muller, F., S. S. Fr0land, M. Hvatum, J. Radl, and P. Brandtzaeg. 4. Delacroix, D. L., C. Dive, J. C. Rambaud, and J. P. Vaerman. 1982. 1991. Both IgA subclasses are reduced in parotid saliva from IgA subclasses in various secretions and in serum. Immunology patients with AIDS. Clin. Exp. Immunol. 83:203-209. 47:383-385. 19. Phelan, J. A., B. R. Salzman, G. H. Friedland, and R. S. Klein. 5. Epstein, J. B., L. H. Kimura, T. W. Menard, E. L. Truelove, and 1987. Oral findings in patients with acquired immune deficiency N. N. Searsall. 1982. Effects of specific antibodies in the interac- syndrome. Oral Surg. 64:50-56. tion between the fungus Candida albicans and human oral mucosa. 20. Reinholdt, J., P. Krogh, and P. Holmstrup. 1987. Degradation of Arch. Oral Biol. 17:469-474. IgAl, IgA2 and S-IgA by Candida and Torulopsis species. Acta 6. Fubara, E. S., and R. Freter. 1973. Protection against enteric Pathol. Microbiol. Immunol. Scand. 95:265-274. bacterial infection by secretory IgA antibodies. J. Immunol. 111: 21. Schiodt, M. 1992. HIV-associated salivary gland disease: a review. 395-403. Oral Surg. Oral Med. Oral Pathol. 73:164-167. 7. Gelderblom, H. R. 1991. Assembly and morphology of HIV: 22. Svandborg-Eden, C., and A. M. Svennerholm. 1978. Secretory potential effect of structure on viral function. AIDS 5:617-638. immunoglobulin A and G antibodies prevent adhesion of Esche- 8. Gough, P. M., D. W. Warnoch, M. D. Richardson, N. J. Mansell, richia coli to human urinary tract epithelial cells. Infect. Immun. and J. M. King. 1984. IgA and IgG antibodies to Candida albicans 22:790-797. in the genital tract secretions of women with or without vaginal 23. Tylenda, C. A., J. Larsen, C.-K. Yeh, H. C. Lane, and P. C. Fox. candidiasis. Sabouraudia 22:265-271. 1989. High levels of oral yeasts in early HIV-1 infection. J. Oral 9. Grey, H. M., C. A. Abel, W. J. Yount, and H. G. Kunkel. 1968. The Pathol. Med. 18:520-524. subclass of human gamma-A-globulins (gA2) which lacks the 24. Vudhichamnong, K., D. M. Walker, and H. C. Ryley. 1982. The disulphide bonds linking heavy and light chains. J. Exp. Med. effect of secretory immunoglobulin A on the in vitro adherence of 128:1233-1236. the yeast Candida albicans to human epithelial cells. Arch. Oral 10. Jackson, S. 1990. Secretory and serum IgA are inversely altered in Biol. 27:617-621. AIDS patients, p. 665. In T. T. MacDonald, S. J. Challacombe, 25. Walker, W. A., M. Wu, K. J. Isselbacher, and K. J. Bloch. 1975. P. W. Bland, C. R. Stokes, R. V. Heatly, and A. M. Mowat (ed.), Intestinal uptake of macromolecules III. Studies on the mecha- Advances in mucosal immunology. Kluwer Academic Publishers, nism by which immunization interferes with antigen uptake. J. Lancaster, United Kingdom. Immunol. 115:854-861. 11. Jeganathan, S., D. Ufomata, J. A. Hobkirk, and L. Ivanyi. 1987. 26. Williams, R. C., and R. J. Gibbons. 1972. Inhibition of bacterial Immunoglobulin Al and A2 subclass of salivary antibodies to adherence by secretory immunoglobulin A: a mechanism of anti- Candida albicans in patients with oral candidosis. Clin. Exp. gen disposal. Science 177:697-699. Immunol. 70:316-321. 27. Wray, D., D. H. Felix, and C. G. Cumming. 1990. Alteration of 12. Kilian, M., J. Mestecky, and M. W. Russell. 1988. Defence humoral responses to Candida in HIV infection. Br. Dent. J. mechanisms involving Fc-dependent functions of immunoglobulin 168:326-329.
You can also read